Background and aim:

Medically managed opioid withdrawal (detox) can increase the risk of subsequent opioid overdose. We assessed the association between mortality following detox and receipt of medications for opioid use disorder (MOUD) and residential treatment after detox.

Design:

Cohort study generated from individually linked public health datasets

Setting:

Massachusetts, USA

Participants:

30,681 opioid detox patients with 61,819 detox episodes between 2012–2014

Measurements:

Treatment categories included no post-detox treatment, MOUD, residential treatment, or both MOUD and residential treatment identified at monthly intervals. We classified treatment exposures in two ways: a) ‘On Treatment’ included any month where a treatment was received, b) ‘With Discontinuation’ individuals were considered exposed through the month following treatment discontinuation. We conducted multivariable Cox proportional hazards analyses and extended Kaplan-Meier estimator cumulative incidence for all-cause and opioid-related mortality for the treatment categories as monthly time varying exposure variables.

Findings:

Twelve months after detox, 41% received MOUD for a median of 3 months, 35% received residential treatment for a median of 2 months, and 13% received both for a median of 5 months. In On Treatment analyses for all-cause mortality, compared with no treatment, adjusted hazard ratios (AHR) were 0.33 (95% CI: 0.26–0.43) for MOUD, 0.63 (95% CI: 0.47–0.84) for residential treatment, and 0.11 (95% CI: 0.03–0.42) for both. In With Discontinuation analyses for all-cause mortality, compared with no treatment, AHRs were 0.51 (95% CI: 0.42–0.62) for MOUD, 0.76 (95% CI: 0.59–0.96) for residential treatment, and 0.20 (95% CI:0.08–0.55) for both. Results were similar for opioid-related overdose mortality.

Conclusions:

Among people who have undergone medically managed opioid withdrawal, receipt of medications for opioid use disorder (MOUD), residential treatment, or the combination of MOUD and residential treatment were associated with substantially reduced mortality compared with no treatment.